Back to EveryPatent.com
| United States Patent |
6,063,643
|
|
Dutta
|
May 16, 2000
|
Surface-emission type light-emitting diode and fabrication process
thereof
Abstract
An n-type GaAs layer as a buffer layer, an n-type (Al.sub.0.7
Ga.sub.0.3).sub.0.5 In.sub.0.5 P layer, an active layer, a p-type
(Al.sub.0.7 Ga.sub.0.3).sub.0.5 In.sub.0.5 P layer, a thin layer of
Al.sub.x Ga.sub.1-x As layer (x.gtoreq.0.9), an A1.sub.0.7 Ga.sub.0.3 As
layer as a current spreading layer and a high doped p-type GaAs cap layer
are sequentially grown on an n-type GaAs layer of a substrate. As the
active layer, an (Al.sub.x Ga.sub.1-x).sub.0.5 P based bulk or
multi-quantum well is employed. As the current spreading layer, an
Al.sub.x Ga.sub.1-x As (x.gtoreq.0.7) is employed. The current spreading
layer is a p-type III-IV compound semiconductor having wider band gap than
a band gap of a material used for forming the active layer, and being
established a lattice matching with the lower p-type cladding layer. After
mesa etching up to the cladding layer, growth of selective oxide is
performed at a part of the AlGaAs layer. BY this, a block layer (selective
oxide of AlGaAs) is formed. By this blocking layer, a light output power
and a coupling efficiency are improved.
| Inventors:
|
Dutta; Achyut Kumar (Tokyo, JP)
|
| Assignee:
|
NEC Corporation (JP)
|
| Appl. No.:
|
272740 |
| Filed:
|
March 8, 1999 |
Foreign Application Priority Data
| Current U.S. Class: |
438/39; 257/E33.027; 257/E33.067; 257/E33.068; 257/E33.07; 438/26; 438/46 |
| Intern'l Class: |
H01L 021/00 |
| Field of Search: |
438/25,26,39,46,47
|
References Cited
U.S. Patent Documents
| 5162878 | Nov., 1992 | Sasagawa et al. | 257/88.
|
| 5181220 | Jan., 1993 | Yagi | 257/92.
|
| 5457328 | Oct., 1995 | Ishimatsu et al. | 257/95.
|
| 5466950 | Nov., 1995 | Sugawara et al. | 257/94.
|
| 5517039 | May., 1996 | Holonyak, Jr. et al. | 257/94.
|
| 5861636 | Jan., 1999 | Dutta et al. | 257/91.
|
Primary Examiner: Picardat; Kevin M.
Attorney, Agent or Firm: Hayes, Soloway, Hennessey, Grossman & Hage, P.C.
Parent Case Text
This is a divisional of application Ser. No. 09/061,622 filed on Apr. 16,
1998 , now U.S. Pat. No. 5,972,731 which is a divisional of U.S.
application Ser. No. 08/706,517 filed Sep. 4, 1996 now U.S. Pat. No.
5,821,569 issued Oct. 13, 1998.
Claims
What is claimed is:
1. A fabrication process for a surface emission type diode having a light
emitting surface comprising the steps of:
sequentially stacking a first conductivity type buffer layer on a first
conductivity type substrate, a first conductivity type cladding layer, an
active layer, a second conductivity type cladding layer, a second
conductivity type thin layer, a second conductivity type current spreading
layer, and a high doped second conductivity type cap layer,
forming a high resistance region in said second conductivity type current
spreading layer, and
forming a molded lens having a selected refraction index, lens diameter,
and distance from said light emitting surface of said surface emitting
type diode to a top at the center of said lens.
2. A fabrication process as claimed in claim 1, and further comprising the
steps of forming a ring-shaped electrode having one or more electrode
rings formed on a light emitting surface side and covering the light
emitting surface other than said ring electrode with an insulative
material having transparence for radiated light.
3. A fabrication process as claimed in claim 1, further comprising the step
of forming a distributed Bragg reflector (DBR) mirror on said first
conductivity type buffer layer.
Description
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates generally to an optical device to be employed
for optical communication system. More specifically, the invention relates
to a high-brightness visible light-emitting diode (LED) to be employed as
high-efficiency light source in a plastic optical fiber (POF) based
optical data-link system. Such LED is also applicable for an outdoor
display device, an automotive indicator and so forth.
2. Description of the Related Art
A light output efficiency of the conventional light-emitting diode (LED) is
mainly dependent on a structure and shape of an electrode to be used for
injecting the current to junction. Conventionally fabricated
surface-emission type light-emitting diode has a quadrangular or
circular-shaped upper electrode located at the center portion thereof. The
area and shape of the upper electrode significantly influence the light
output efficiency. For achieving higher light output efficiency, it is
required to make the area of the electrode smaller and the area of a
light-emitting surface wider. In the conventional LED, while employment of
the quadrangular or circular-shaped electrode positioned at the center may
facilitate fabrication, but it may cause disruption of gaussian beam
pattern of an output beam. Especially, it may make the beam having the
peak of the emission pattern at the circumferential edge portion of the
electrode. The emission pattern attenuate according to increasing of
distance from the electrode. This is caused by the fact that the current
is concentrated at the center portion and not distributed at the position,
distanced from the electrode. A beam intensity to be attained is directly
proportional to a current density thereat. The conventional LED having
wide beam pattern as set forth above cannot be employed in an optical
data-link system. That is, even when a large core fiber is employed, the
coupling efficiency is held low due to emission pattern.
One of typical examples of the conventional surface emission type LED is
shown in FIGS. 1A, 1B and 2. FIGS. 1A and 1B show sections along an upper
surface of an LED and an approximated near-field-emission. In FIGS. 1A and
1B, an n-type GaAs buffer layer 2, an n-type (Al.sub.x
Ga.sub.1-x)In.sub.1-x P layer 3, an active layer 4, a p-type (Al.sub.x
Ga.sub.1-x).sub.0.5 In.sub.0.5 P layer 5, a current spreading layer 7 and
a p-type GaAs cap layer 8 are grown sequentially on an n-type GaAs
substrate 1. For a p-type contact, quadrangular, circular or cross shaped
electrode 28 is employed in the prior art. As the current spreading layer
7, a thick Al.sub.0.7 Ga.sub.0.3 As layer is normally employed.
In FIG. 2, an approximated near-field-pattern of a light intensity 30 for
LED having a quadrangular or circular shaped electrode located at the
center, is shown. As shown, the light intensity 30 is maximum around the
peripheral edge portion of the electrode and becomes lower at distance
position from the electrode. This represents crowding of the current at
the center and not distributing far away for the electrode. In this case,
it is impossible to enhance the output efficiency as expected. In order to
avoid crowding of current around the electrode, a blocking layer 31 is
typically employed in front of the AlGaAs layer (current spreading layer)
7. FIG. 3 shows another example of the conventional LED having the
blocking layer 31. Requirement of regrowth makes production cost of the
LED higher. Also, darkness at the center portion in the case of circular
electrode makes the coupling efficiency lowers even with a large core
fiber. It is highly desirable to provide an LED structure which can
provide high coupling efficiency and is suitable for mass scale
production.
III-V semiconductor based visible LED (wavelength is in a range from 580 to
670 nm) which has quadrangular/circular shaped electrode at the center,
have been disclosed in various publications and patent publications. In
every case, quadrangular or circular shaped electrode provided at the
center is used as upper contact for injecting the current. One of typical
examples can be found in the paper of Sugawara et al., "Japan Journal of
Applied Physics, Part 1, Vol. 31, No. 8, pp 2446 to 2451, 1992". In this
report, a cross shaped electrode along with blocking layer were used in
the surface-emission type LED.
However, the light output efficiency is still beyond its practical
application. To enhance current spreading outside the contact, a thick
window layer having low resistivity is required. Even with such thick
window layer, the current spreading outside the contact is limited to a
certain level. Thus, the light intensity is much lower than that required
for practical application. However, in application, such as short distance
data-link system, especially based on POF, employment of the conventional
LED exhibits low coupling efficiency and thus is impractical in a POF
based communication system. For POF based data link application, it would
be highly desirable to design the LED which offers not only high
brightness but also high coupling efficiency. In addition, it is highly
desirable to reduce a fabrication cost as low as possible in order to
contribute for lowering of a cost for the system.
Japanese Unexamined Patent Publication (Kokai) No. Heisei 2-174272
discloses a high brightness LED. In the disclosed art, the brightness of
the LED is enhanced employing n-p-n-p structure under the contact, which
assists for spreading the current outside the contact. The fabrication
process of the conventional LED as disclosed in the above mentioned
publication includes a first step of sequentially growing of an N-type
junction layer, a P-type light emitting layer and an N-type transparent
layer on an N-type semiconductor substrate, a second step of performing
photo-etching for the N-type transparent layer, a third step of performing
diffusion of Zn for forming a current spreading region, a fourth step of
performing photo-etching for isolation of elements and polishing of the
lower portion of the element, and fifth step of forming electrode on upper
and lower portions of the element. The main drawback of the disclosed
invention is that prior to form p-type contact, a mesa structure reaching
to the active region has to be formed in a light emitting surface to
define a current path. Also, a dopant, such as Zn, has to be diffused at
high temperature environment. The post-growth high temperature treatment
for Zn diffusion influences on performance characteristics due to
degradation of the active region. As the light emitting surface with high
p-doping is used, a large fraction of light (depending upon the light
energy) can be absorbed in the diffused layer. In fabrication of this type
of LED, several times of processes are required to cause high fabrication
cost of the LED. In addition, difficulty may be encountered in fabrication
of high speed LED for high capacitance caused by wide contact area.
Japanese Unexamined Patent Publication No. Heisei 3-20 190287 discloses a
high brightness LED. In the disclosed conventional LED, a plurality of
light emitting regions of mesa structure are arranged on a direction. Each
of the light emitting regions is formed so that edges of the light
emitting region perpendicular to the arranging direction thereof become
forward mesa structure and remaining edges parallel to the arranging
direction become reverse mesa structure. And an electrode is lead from
only one of edges having forward mesa structure. In such LED array, for
facilitating wire bonding, the contact was made on the upper layer
following the mesa formation. In this case, the shape of the contact is
varied to achieve high light output. The disclosed prior art is completely
related to the LED array, and since the mesa structure is required to be
formed prior to the electrode formation, there should have difficulties in
formation of the electrode unless a thick contact is formed. This is not
only results in increasing of the fabrication cost of the LED, but also
impractical for application in fabrication of single LED.
Japanese Unexamined Patent Publication No. Heisei 5-211345 discloses an
LED. In the disclosed prior art, between a p-type clad layer and a cap
layer, an n-type current blocking layer is provided. In conjunction
therewith, a portion to be the peripheral edge of the upper electrode in
the current blocking layer is converted into p-type by diffusion or ion
implantation of a p-type impurity through a portion of a light extraction
surfaces on the cap layer where an upper electrode is not mounted. In such
prior art, the current blocking layer, the type of which is opposite type
(p or n type) of compound semiconductor, and diffusion at high temperature
are used for fabrication of the LED. In addition, the configuration of the
upper electrode of the LED is impractical to be employed as a light source
in the optical data-link system for low coupling efficiency. Furthermore,
high temperature treatment degrades the optical characteristics due to
dopant diffusion and also crystal defect.
The Japanese Unexamined Patent Publication No. Heisei 4-174567 discloses a
high brightness LED array to be employed in a printer. The disclosed
conventional surface emission type light emitting diode array has a
plurality of surface emission type light emitting diodes extracting a
light generated in a plurality of active layers provided on a common
substrate, through light extracting surfaces formed at opposite side of
the substrate. Also, a light reflector layer of semiconductor multiple
layers are provided. The light reflector layers are respectively provided
in between active layer of surface emission type light emitting diode and
the common substrate for reflecting the light by light wave interference.
In such prior art, a bottom distributed bragg reflector layer (DBR) is
used for reflecting the light back from the substrate. In this case, the
same idea for fabricating the surface emission laser is implemented. For
the DBR, pairs of GaAs/AlAs is used for reflecting a 880 nm wavelength
light toward the substrate. The types and number of pairs to be used in
the DBR should be dependent on the output emission wavelength. Also, the
same DBR cannot be used in the visible LED where 600 to 650 nm wavelength
are concerned.
Namely, the disclosed LED array can be used in the printer but cannot be
employed in the data link system. This is because the configuration of the
upper electrode is the same as that of the conventional LED as set forth
above. The coupling efficiency with a fiber becomes quite low as such LED
is employed.
Japanese Unexamined Patent Publication No. Heisei 4-259263 discloses a
visible LED employing InAlGaP. The disclosed conventional semiconductor
light emitting element is fabricated by sequentially growing n-type clad
layer formed of InGaAlP type material, active layer and p-type clad layer,
on an n-type GaAs substrate to form a double heterostructure portion. A
p-type intermediate band gap layer is formed on the double heterostructure
portion, and a p-type contact layer is selectively formed on the
intermediate band gap layer. The semiconductor light emitting element is
consisted of an active layer formed with an ordered layer having natural
super lattice, and the intermediate band gap layer 15 formed with a
non-ordered layer, in which the natural super lattice is extinguished by
Zn diffusion. By making the band gap of the intermediate band gap layer
greater than the band gap of the active layer, the light from the light
emitting region can be extracted without blocking by the electrode at the
light extracting side. In the disclosed prior art, an additional layer of
InGaP, band gap wider than the layer used in active region is used in
order to avoid Zn diffusion into the active region. No additional layer
for current spreading is used. As set forth above, this type of the
visible LED with centrally located circular shaped electrode can not be
used in the short distance data link system.
As explained in the above, the LED structure so far proposed, has light
output and coupling efficiency, not enough for using in the short distance
data link system, especially where POF based system are concerned. This is
because, conventional LED has low output power and also low coupling
efficiency with the POF. Therefore, it would be highly desirable to design
a high brightness visible LED which could be fabricated in low cost and
suitable for mass scale production.
SUMMARY OF THE INVENTION
It is an object of the present invention to provide a surface emission type
light-emitting diode and production process therefor, which can offer high
output, high coupling efficiency with POF, and high brightness with
uniformly spreading current outside a contact.
The problems set forth above may be alleviated for its potential
application in POF or silica based data link system or in any other
display system.
These problems can be alleviated by uniformly distributing current on the
emitting surface, and this can be possible by preventing current from
spreading toward bonding region. If current is not allowed to spread
toward the bonding region, current could be injected only to the junction
of the emitting surface, circular shaped emitting surface is used to match
the outgoing beam shape with the fiber core, resulting in increased
coupling efficiency. Uniform emission pattern is also achieved by
optimizing (selection thereof depends on dimension of fiber) the emitting
surface diameter, the selection of which also dependent on fiber
dimension. The LEDs proposed in the invention can also be useful in
display system.
BRIEF DESCRIPTION OF THE DRAWINGS
The present invention will be understood more fully from the detailed
description given herebelow and from the accompanying drawings of the
preferred embodiment of the invention, which, however, should not be taken
to be limitative to the present invention, but are for explanation and
understanding only.
In the drawings:
FIGS. 1A and 1B are schematic diagrams of the conventional visible LED;
FIG. 2 is an illustration showing approximated near field pattern of the
LED of FIGS. 1A and 1B;
FIG. 3 is a schematic diagram showing a structure of the conventional LED;
FIGS. 4A and 4B are illustrations showing the first embodiment of an LED
according to the present invention;
FIG. 5 is a graph showing a result of calculation representing dependency
of light output efficiency on a dimension of light emitting surface;
FIG. 6 is an illustration showing the second embodiment of the LED having a
selective oxide layer as a DBR and a block layer;
FIGS. 7A and 7B are illustrations showing a DBR structure employed in the
second, fourth, sixth, eighth and twelfth embodiment of the LEDs together
with reflection index characteristics;
FIG. 8 is an illustration showing the third embodiment of the LED having an
ion implantation blocking layer;
FIG. 9 is an illustration showing the fourth embodiment of the LED having
the ion implantation block layer and a bottomed DBR;
FIG. 10 is an illustration showing the fifth embodiment of the LED having
mesa etched for avoiding spreading of a current at lower side of the
bonding portion;
FIG. 11 is an illustration showing mesa etched the sixth embodiment of the
LED having the bottomed DBR;
FIG. 12 is an illustration showing the seventh embodiment of the LED having
a blocking layer for improving light output efficiency;
FIG. 13 is an illustration showing the eighth embodiment of the LED having
a bottom electrode;
FIG. 14 is an illustration showing the ninth embodiment of the LED having a
ring shaped electrode;
FIG. 15 is an illustration showing the tenth embodiment of the LED having
the ring shaped electrode and the bottomed DBR;
FIGS. 16A and 16B are illustrations showing a mold structure for the
eleventh embodiment of the LED together with POF;
FIG. 17 is a graph showing a result of calculation showing dependencies of
coupling efficiency on lens distance and fiber distance; and
FIG. 18 is a graph showing a result of calculation showing dependencies of
coupling efficiency on lens distance and fiber distance.
DESCRIPTION OF THE PREFERRED EMBODIMENT
The preferred embodiments of the present invention will be discussed in
detail with reference to the accompanying drawings. In the following
description, numerous specific details are set forth in order to provide a
thorough understanding of the present invention. It will be obvious,
however, to those skilled in the art that the present invention may be
practiced without these specific details. In other instance, well-known
structures are not shown in detail in order to unnecessary obscure the
present invention.
FIGS. 4A and 4B are explanatory illustration showing a structure of the
first embodiment of a visible LED according to the present invention. FIG.
4A is a section and FIG. 4B is a top plan view. As shown, an n-type GaAs
layer 2 as a buffer layer, an n-type (Al.sub.0.7 Ga.sub.0.3).sub.0.5
In.sub.0.5 P layer 3, an active layer 4, a p-type (Al.sub.0.7
Ga.sub.0.3).sub.0.5 In.sub.0.5 P layer 5, a thin layer of Al.sub.x
Ga.sub.1-x As layer 6 (x.gtoreq.0.9), a current spreading layer Al.sub.0.7
Ga.sub.0.3 As layer 7, a high doped p-type GaAs cap layer 8 are
sequentially grown on a substrate n-type GaAs layer 1. As the active layer
4, a bulk or multiquantum well of (Al.sub.x Ga.sub.1-x).sub.0.5 In.sub.0.5
P based material of a desired wavelength can be used. In the preferred
embodiment, Al.sub.x Ga.sub.1-x As with x content of .gtoreq.0.7 is used
as the current spreading layer 7. However, it also covers all p-type III-V
compound semiconductor having a band gap wider than that of used as active
layer and also have the lattice matching with its underlying p-type
cladding layer.
Since high doping is necessary in the current spreading layer 7 and the cap
layer 8, dopant diffusion into the active layer may deteriorate optical
characteristics of the LED during epitaxial growth. In order to prevent
the dopant from diffusing into the active layer, a spacer layer 5(1)
having a thickness depending on thicknesses of the current spreading layer
7 and the cap layer 8, is formed following formation of the active layer
4. As the spacer layer, the same material as the p-type material with low
doped used as the p-type clapping layer 6 may be used.
After mesa etching reaching up to p-type cladding layer 5(2), selective
oxide growth in part of AlGaAs 6(2) is performed. This can be done by
heating a sample at a temperature over 400.degree. C. in steam atmosphere
which can be realized by water bubbling. As the Al contents in AlGaAs is
more than 0.9, selective oxide growth can be done in the layer 6(2). The
selective oxide growth should be selected in such a way that the opening
portion (unoxide portion) 6(1) is equal to the diameter of a light
emitting surface 12.
A current is injected to a junction through the opening portion 6(1). In
the preferred embodiment, a thin layer of AlGaAs 6 is used. However, any
type of III-V compound semiconductor which has crystal matching with the
cladding layer 5 and has capability of selective oxidation (e.g. AlAs) may
be used for forming the thin layer 6. After epitaxial growth, a silicon
nitride or silicon oxide 9 in a thickness greater than or equal to 150 nm
is deposited at a temperature lower than or equal to 500.degree. C.
Subsequently, an opening window through the dielectric layer 9 for the
light emitting surface 12 is formed. Then, a p-type contact electrode
10(10(1) and 10(2)) is formed using a lift-off process.
In the LED, as a current density is higher in comparison with a current
density of the optical device like a laser diode, the contact resistance
mainly induced due to the p-type electrode, should be as low as 1 ohm for
long time reliability. Therefore, as the p-type contact, more reliable
metallization has to be selected in order to avoid any spike formation in
the GaAs cap layer. For the p-type contact, Au:Zn or Ti/Pt/Au or Ni/Zn/Au
can be used with much higher reliability in comparison with metallization
of the normal kind, such as Cr:Au.
For using the LED in the data link system, the optical coupling efficiency
with the fiber is more concerned. Therefore, the emission surface should
be designed so that a maximum coupling efficiency can be achieved. The
coupling efficiency is mainly depend on several factors: the type of the
fiber to be used and the emission surface of the LED to be used. In case
of the LED, the light emitting surface should be designed in such a way
that the more light is concentrated at the center region rather than outer
side. The coupling efficiency is much more dependent on selection of the
emitting surface diameter (namely the diameter of the upper surface). The
coupling efficiency can be maximized by optimizing the diameter of the
upper electrode which corresponds the light emitting surface.
As set forth above, when the conventional electrode which is mainly
provided at the center portion, is employed, the output light is broadened
at the outer side of the LED to lower the coupling efficiency. If a ring
shaped electrode and a current blocking under the bonding region 16 are
employed, the light may be concentrated at the center to achieve higher
coupling efficiency in comparison with that using the conventional LED.
The design of the circular electrode along with the circular light
emitting surface is also important factor for enhancing the coupling
efficiency and its brightness. In order to enhance the output power, the
diameter of the electrode (diameter of the light emitting surface) should
be optimized.
FIG. 5 shows a dependency of the light emitting surface diameter on the
light output efficiency as a function of current spreading layer (AlGaAs)
thickness. When the thicker current spreading layer is employed, higher
brightness can be attained with a given fixed light emitting surface area.
However, when the light emitting surface is widened, the light output
efficiency becomes closer to that of the LED having thicker current
spreading layer. In order to achieve the maximum coupling efficiency
(.gtoreq.50%) with the fiber, the diameter of the light emitting surface
should be designed in such a way that ratio of the fiber core diameter to
the light emitting surface is greater than or equal to 5. It has been
found that if the step index (SI) POF fiber with the core size of 0.98 mm
and numerical aperture (NA) of 0.5 are used, the ring diameter (light
emitting surface) to be used should be as low as 100 .mu.m with achieving
the coupling efficiency higher than or equal to 80%.
After forming the p-type contact, the Au 10(2) is plated on the p-type
contact or partially on the portion where a wire bonding is effected, for
facilitating wire bonding. This is followed by covering the light emitting
surface 12 by a passivation layer 13. For the passivation layer 13.
alumina (A1.sub.2 O.sub.3), silicon oxide (SiO.sub.2) or silicon nitride
(SiN.sub.x) can be used. The passivation layer 13 has to be deposited at a
room temperature or at a temperature lower than or equal to 200.degree. C.
The passivation layer 13 prevents the light emitting surface 12 from
absorbing water molecular or oxygen from the environment and thus makes
device reliable for operation over a long period. After formation of the
passivation layer 13, the back side of the substrate is polished to around
150 .mu.m. Then, the n-type contact 11 is formed thereon. For the n-type
contact 11, Au:Ge/Ni/Au can be used. After formation of the n-type contact
11 on the back side of the polished substrate, each LED is scrubbed for
the packaging. The use of the blocking layer (in this case selective
AlGaAs oxide) and simple structure of LED is effective not only for
improving light output power and coupling efficiency, but also for
minimizing fabrication cost of the LED.
FIG. 6 shows an explanatory diagram of the second embodiment of the LED, in
which like parts to the foregoing first embodiment will be indicated by
the like reference numerals in the first embodiment for simplification of
disclosure with avoiding redundant discussion. In the second embodiment,
the distributed bragg reflector (DBR) mirror 14 is grown on the buffer
layer prior to growing of the cladding layer 3, the active layer 4 and the
current spreading layer 7. The DBR to be used as the mirror 14, consists
of the semiconductor pairs of 14(1) and 14(2) having high refraction index
difference. Number and the type of the pairs also determine the level of
the reflection of the DBR. The semiconductor layer to be used as the DBR
should also have low absorbency for the emitted light.
FIGS. 7A and 7B show explanatory example of the DBR along with its
approximated reflecting characteristics. As set forth above, number of
pairs as well as their type to be used in the pairs are dependent on the
output light wavelength. In the optical device, especially the LED, as
emission spectra covers the broad wavelength, the DBR to be used should
have high reflectivity in broad wavelength range. As shown in FIG. 7B, the
maximum reflectivity in broad wavelength, ranging from .lambda..sub.1 to
.lambda..sub.2 should be achieved. This would assist to achieve the
maximum reflectivity over the wafer area, while there will be shifting of
the DBR reflectivity due to non-uniformity of the thickness, if any. In
designing LED of 650 nm, AlA.sub.x, Al.sub.x Ga.sub.1-x A(x.gtoreq.0.45),
(Al.sub.x Ga.sub.1-x).sub.0.5 In.sub.0.5 P (x.gtoreq.0.45), Ga.sub.0.5
In.sub.0.5 P or GaAs can be used as DBR materials. The single DBR or
discrete DBR can be used to achieve the broad wavelength ranges.
Table 1 and table 2 summarize pairs type and the DBR characteristics usable
in designing visible LED of wavelength 650 nm.
TABLE 1
______________________________________
Summarizing the DBR type and its
characteristics usable in the visible LED
Wavelength
Pair Type Reflectivityf
Ranges
(DBR type) (%)irs
.sub.1 -.sub.2 (nm)
______________________________________
AlAs/ .gtoreq.20
99.9 620-685
Al.sub.0.5 Ga.sub.0.5 As
Al.sub.0.5 In.sub.0.5 P/
.gtoreq.15
99.9 630-670
(Al.sub.0.5 Ga.sub.0.5).sub.0.5 In.sub.0.5 P
(AL.sub.0.5 In.sub.0.5 P/
.gtoreq.25
99.9 620-680
Ga.sub.0.5 In.sub.0.5 P
AlAs/ 99.9 615-685
(Al.sub.0.5 Ga.sub.0.5).sub.0.5 In.sub.0.5 P
AlAs/ 90.0 610-685
GaAs
(Al.sub.0.95 Ga.sub.0.05).sub.0.5 In.sub.0.5 P/
.gtoreq.30
99 620-680
(Al.sub.0.02 Ga.sub.0.98).sub.0.5 In.sub.0.5 P
Al.sub.0.5 Ga.sub.0.5 As/
.gtoreq.30
>95 625-675
(Al.sub.0.95 Ga.sub.0.05).sub.0.5 In.sub.0.5 P
Al.sub.0.5 Ga.sub.0.5 As/
.gtoreq.15
>82 600-700
GaAs
Al.sub.0.5 In.sub.0.5 P
.gtoreq.15
>90 610-685
GaAs
(Al.sub.0.5 Ga.sub.0.5).sub.0.5 In.sub.0.5 P/
.gtoreq.15
>95 625-685
(Al.sub.0.95 Ga.sub.0.05).sub.0.5 In.sub.0.5 P
______________________________________
TABLE 2
______________________________________
Summarizing the DBR type and its
characteristics usable in the visible LED
Wavelength
Pair Type Reflectivity
Ranges
(DBR type) (%)
.sub.1 -.sub.2 (nm)
______________________________________
AlAs/Al.sub.0.5 Ga.sub.0.5 As +
10 + 10 99.9 610-690
Al.sub.0.5 Ga.sub.0.5 As/
(Al.sub.0.95 Ga.sub.0.05).sub.0.5 In.sub.0.5 P
GaAs/AlAs + 93.0 606-69810
(Al.sub.0.95 Ga.sub.0.05).sub.0.5 In.sub.0.5 P/
Ga.sub.0.5 In.sub.0.5 P
______________________________________
In designing DBR, formation of the graded junction to minimize its
resistance is also important similarly to the reflection characteristics.
For this, super lattice or graded junction matching with pairs is to be
used. After formation of the DBR 14, other layers, such as the n-type
cladding layer 4, the active layer 4, p-type cladding layer 5, the current
spreading layer 7 and the p-type GaAs cap layer 8 are sequentially grown
in the single chamber. Other processes following this step are the same as
those described in the first embodiment. Therefore, discussed for the
subsequent steps is neglected for maintaining the disclosure simple enough
for facilitating clear understanding of the present invention.
FIG. 8 is an explanatory diagram showing the proposed structure of the
third embodiment of LED, in which the like parts are indicated by the like
reference numerals in the first embodiment so that the repeated
explanations are omitted, In the third embodiment, mesa etching is not
required. Before deposition of dielectric layer 9, such as silicon oxide
or silicon nitride, proton implantation 15 is performed after covering the
light emitting surface by the thick resist or any other types of materials
which can be etched selectively. The proton implantation is done to make
the high resistive region for avoiding the current from spreading toward
the bonding region 16. The selective oxide layer 6(2) used in the first
embodiment is not required in the shown embodiment. However, it is also
covers the LED structure where the ion implantation along with the
selective oxide layer is also used. The use of ion implantation along with
the blocking layer can help to enhance light output and coupling
efficiency. The processes following to this are already explained in the
first embodiment, so that the repeated explanation thereof is omitted.
FIG. 9 is an explanatory diagram showing the proposed structure of the
fourth embodiment of the LED according to the present invention. In the
shown embodiment, the like parts are represented by the like reference
numerals in the first, second and third embodiments so that the repeated
explanation can be omitted. In the fourth embodiment, after formation of
the DBR 14, the n-type cladding layer 3, the active layer 4 and the p-type
cladding layer 5 are sequentially grown. The current blocking region is
formed by way of proton implantation 15. The processes following this has
been discussed already in the second embodiment. The discussion for the
following step is neglected for avoiding redundant discussion.
FIG. 10 is the explanatory diagram showing the proposed structure of the
fifth embodiment of the LED according to the present invention. It should
be noted that the like parts are represented by like reference numerals in
the first embodiment so that the repeated explanation can be omitted. In
the fifth embodiment, the thin layer 6 used in the first embodiment is not
used. In the shown embodiment, the current is injected to the junction
using the mesa structure having a slightly wider area than that of the
light emitting surface. The bonding portion 17 is located outside the mesa
structure. After mesa etching, the dielectric layer is deposited. Other
processes following this step are the same as those described in the first
embodiment. Therefore, discussed for the subsequent steps is neglected for
maintaining the disclosure simple enough for facilitating clear
understanding of the present invention. Owing to simple structure, this
LED can be fabricated in low cost and thus is suitable for mass scale
production.
FIG. 11 is an explanatory diagram showing the proposed structure of the
sixth embodiment of the LED according to the present invention. It should
be noted that the like parts are represented by like reference numerals in
the first, second and fifth embodiments so that the repeated explanation
can be omitted. In the sixth embodiment, following the process step of
growing the DBR 14, the n-type cladding layer 3, the active; layer 4, the
p-type cladding layer 5, the current spreading layer 7 and the cap layer 8
are subsequently grown on the DBR. Other processes following this step are
the same as those described in the first embodiment. Therefore, discussed
for the subsequent steps is neglected for maintaining the disclosure
simple enough for facilitating clear understanding of the present
invention.
FIG. 12 shows an explanatory diagram illustrating the proposed structure of
the seventh embodiment of the LED according to the present invention. It
should be noted that the like parts are represented by like reference
numerals in the first embodiment so that the repeated explanation can be
omitted. In the shown embodiment, the thin layer 6 and the mesa etching
area are required. Also, in the shown embodiment, for fabricating the LED,
two steps of epitaxial growth are required. After growing the p-type
cladding layer 5, a blocking layer 18 is to be grown and subsequently, a
pattern is formed. The blocking layer 18 assists prevention of current
from concentrating under the bonding region. As the blocking layer 18, the
same n-type material, used as the cladding layer 3, may be used. After the
patterning of blocking layer 18 is performed, the current spreading layer
7 and a high dope cap layer of GaAs 8 for the p-type contact are grown
sequentially. Other processes following this step are the same as those
described in the first embodiment. Therefore, discussed for the subsequent
steps is neglected for maintaining the disclosure simple enough for
facilitating clear understanding of the present invention. The use of the
blocking layer 18 assists to inject current into the junction without
spreading toward the bonding region and to enhance the light output
efficiency.
FIG. 13 is the explanatory diagram illustrating the proposed structure of
the eighth embodiment of the LED. It should be noted that the like parts
are represented by like reference numerals in the first, second and
seventh embodiments so that the repeated explanation can be omitted. After
formation of the DBR 14, the n-type cladding layer 3, the active layer 4,
the p-type cladding layer 5 and the blocking layer 18 are grown
sequentially. Other processes following this step are the same as those
described in the second embodiment. Therefore, discussed for the
subsequent steps is neglected for maintaining the disclosure simple enough
for facilitating clear understanding of the present invention.
FIG. 14 is the explanatory diagram illustrating the proposed structure of
the ninth embodiment of the LED according to the present invention. It
should be noted that the like parts are represented by like reference
numerals in the first embodiment so that the repeated explanation can be
omitted. The thin layer 6, mesa etching, the extra passivation layer 13
used in the first embodiment, are not required. In the ninth embodiment, a
thin layer of GaAs 19 in a thickness of 100 .ANG.A is to be used to form
the p-type contact and also for avoiding the absorption of the emitted
light. In addition, a ring (single or more than single) shaped top
electrode is used to spread current uniformly in the spreading layer.
Following to deposition of the dielectric material 9, a ring shaped window
is opened through the dielectric material. Then, evaporation of the Au:Zn,
Ni/Zn/Au or Cr:Au is carried out for the p-type contact. By the lift off,
the ring electrode on the light emitting surface 20 is formed.
Subsequently, Cr:Au evaporation on the bonding region 21 along with Au
plating 22 covering entire electrode portion, to produce the LED. As the
dielectric material is used for the purpose of bonding portion and also
for passivation layer, it should be selected in such that no absorption
occurs for the emitted light. For example, for 650 nm LED, silicon oxide
or alumina may be used as dielectric material 19. Since only a few process
steps are required for fabrication of the LED, the cost for fabrication of
the chip can be minimized. It should be noted that while the shown
embodiment forms the high resistance region in the current spreading layer
by ion implantation, it may be an oxide blocking layer or an insulative
blocking layer.
FIG. 15 is the explanatory diagram showing the mold structure for the tenth
embodiment of the LED according to the invention. It should be noted that
the like parts are represented by like reference numerals in the first,
second and ninth embodiments so that the repeated explanation can be
omitted. In the tenth embodiment, after formation of the DBR 14, the
n-type cladding layer 3, the active layer 4, the p-type cladding layer,
the current spreading layer and a thin GaAs layer are grown sequentially.
Other processes following this step are the same as those described in the
second embodiment. Therefore, discussed for the subsequent steps is
neglected for maintaining the disclosure simple enough for facilitating
clear understanding of the present invention.
FIGS. 16A and 16B are the explanatory diagrams illustrating the mold
structure for the eleventh embodiment of the LED according to the present
invention. In the eleventh embodiment, after bonding the LED on a lead
frame 24, molding 25 is performed for increasing the coupling efficiency.
A lens 26 is formed during molding. For improving the coupling efficiency
and also keeping the packaging cost as low as possible, the lens 26 to be
used is also made from the same type of molding material. In POF 27 based
data link application, radius (R) of the lens should be .ltoreq.0.6 mm and
distance of the LED from the center of the lens 26(c) should be as low as
0.75 mm. The material to be used for molding should have refractive index
n.ltoreq.1.60, and it should be made at a temperature less than
200.degree. C.
FIGS. 17 and 18 show result of evaluation for the molded lens as shown in
FIG. 16. It is found that the use of the LED 23 with the mold 25 in the
step index (SI) POF 27 having the numerical aperture (NA) of 0.5 and core
diameter of 0.98 mm, can increase the coupling efficiency more than 80%.
The position (X.sub.1) of the LED 23 with respect to the lens center
26(c), the lens distance Z.sub.1 and fiber distance z.sub.2 play an
important role on the coupling efficiency. In order to achieve the
coupling efficiency higher than or equal to 90%, the minimum Z.sub.1
should be 0.7 mm.
In the preferred embodiment set forth above, the LED implementable in the
POF based optical system is explained. The Si POF was used for showing the
practicability of the proposed LED 23 as the optical source in the short
distance data link system. However, it can be also used as an optical
source in all types of POF or conventional silica fiber based system.
Furthermore, the proposed LED is applicable in the outdoor application.
The LED structure described in the preferred embodiment is also applicable
for the LED of wavelength ranging from 0.5 to 1.6 pm.
The foregoing description of the preferred embodiments of the present
invention has been presented for the purpose of illustration and
description, and is not intended to limit the invention to the precise
form disclosed. It has been chosen and described in order to best explain
the principle of the invention and these can be utilized in various
embodiments and with various modifications as are suited to the particular
use contemplated.
In the present invention, the visible LED with the blocking layer under he
bonding region exhibited higher output power as compared with the
conventional one ever developed. It also exhibited coupling efficiency
over 80% with the POF. The use of circular electrode along with circular
light emitting surface assists for spreading the current uniformly outside
the contact, making the LED high brightness and increasing the coupling
efficiency higher than the LED ever fabricated.
Top